Magnetic measurements of neuronal activity provide valuable information about brain function, allowing identification of sources and characterization of system dynamics. Previously, this information was recorded with low spatial resolution at the scale of the whole brain. Emerging new technology based on ultra-sensitive Atomic magnetometers, allows micro-scale neural systems to be probed. Recently, NV-diamond magnetometers attracted considerable interest due to prospects for high sensitivity and resolution of the neuronal magnetic field. Microscopic magnetic field imaging at the level of a few neurons is a novel and potentially revolutionary direction for functional neuroimaging. Magnetic field mapping provides direct information on functionally significant processes in neurons, potentially with better certainty and resolution than other techniques Multi-scale modeling of dynamic neuronal networks is essential for understanding how the human brain works. Such modeling can fill the gaps in inherently limited experimental data, allowing us to improve modeling assumptions. More realistic system models can be tested with proposed experiments. Large-scale neuronal network simulations, currently based on simplified neuron models, will be coupled with anatomically realistic magnetic field calculations to predict magnetic fields of neuronal systems at multiple scales. The models can improve our understanding of the genesis of magnetoencephalography (MEG) and may suggest useful strategies to enhance other methods based on magnetic measurements, such as imaging of neuronal current with MRI. Our multi-faceted collaboration, linking the technology of ultra-sensitive magnetic field measurements based on atomic and NV-diamond magnetometers, with expertise in physiological measurements and large-scale neural network imaging, coupled with detailed magnetic field calculations, enables this challenging but rewarding project.